[FFmpeg-devel] Re: [PATCH] rvv: add optimized h264 intra prediction functions (PR #20494)

[Tristan Matthews via ffmpeg-devel <ffmpeg-devel@ffmpeg.org>]

On Thu, Sep 11, 2025 at 1:29 PM tmatth via ffmpeg-devel
<ffmpeg-devel@ffmpeg.org> wrote:
>
> PR #20494 opened by tmatth
>
> +endfunc
> diff --git a/tests/checkasm/h264pred.c b/tests/checkasm/h264pred.c
> index 53e1cdb219..7254854534 100644
> --- a/tests/checkasm/h264pred.c
> +++ b/tests/checkasm/h264pred.c
> @@ -172,8 +172,15 @@ static void check_pred8x8(H264PredContext *h, uint8_t *buf0, uint8_t *buf1,
>              randomize_buffers();
>              call_ref(src0, 24*SIZEOF_PIXEL);
>              call_new(src1, 24*SIZEOF_PIXEL);
> -            if (memcmp(buf0, buf1, BUF_SIZE))
> +            if (memcmp(buf0, buf1, BUF_SIZE)) {
> +                fprintf(stderr, "REF:\n");
> +                for (int i = 0; i < BUF_SIZE; i++)
> +                    fprintf(stderr, "%d\t", buf0[i]);
> +                fprintf(stderr, "\nASM:\n");
> +                for (int i = 0; i < BUF_SIZE; i++)
> +                    fprintf(stderr, "%d\t", buf1[i]);
>                  fail();
> +            }
>              bench_new(src1, 24*SIZEOF_PIXEL);
>          }
>      }
> --
> 2.49.1
>

This debug code wasn't supposed to be here (in fact I didn't expect
this WIP/draft PR to get mailed out), please see the latest version at
https://code.ffmpeg.org/FFmpeg/FFmpeg/pulls/20494

Best,
Tristan
_______________________________________________
ffmpeg-devel mailing list -- ffmpeg-devel@ffmpeg.org
To unsubscribe send an email to ffmpeg-devel-leave@ffmpeg.org